GCeasy Tool User Review

4.8

6831 Reviews

Reviews written by the users after using the tool

5 star

86.4%

4 star

9.2%

3 star

1.8%

2 star

0.9%

1 star

1.7%

User Comments

2021-09-28

Extraordinary

By sudhakarpft45

2021-09-27

celar, right, this is what i want!

2021-09-23

Awesome, its nice a quick way to understand GC logs.

By army.vinay

2021-09-22

.

2021-09-21

very good

2021-09-15

good

2021-09-11

It's a wonderful tool to analyze GC logs. I love it!

2021-09-03

good

2021-09-02

aaa

2021-09-01

perfect

By igor.kstn

2021-08-27

kbu\\

2021-08-26

great

2021-08-21

helps me alot

2021-08-21

very good!

2021-08-20

Wonderful!

2021-08-13

real np!!

2021-08-05

a

2021-08-04

I tipped you off a more than a year ago about conseuctive full GCs... I might phrase them in the tooltip as: "Consecutive full GCs indicate the application ma...y be trying to use more memory than it's allocated heap can accomidate. At its most severe, this cause memory errors or hang the application."

I like that you specifically report max time for humongous allocations, and indicate min/max times for various stop-the-world events... on the graph, while GC duration may be interesting, what I'd really care to see is world-stop duration. It's one thing if a single GC event that blocks one thread takes 2 or 10 or 20 seconds... It's another if the whole world stops for half a second, multiple times a day.

That being said... this log is for a server providing an interactive web-ui. It indicates a 513 max duration stop-the-world event. If that's common-place, it's a thing that I need to look into. If that's a one-time event, that's essentially a non-event for me.

Sure, I can... sort of, eke out how much of a concern those events might be from the average/stddev you report, but I know the data is going to carry a heavy positive/right-tailed skew to it's population, stddev is >nice< but but in context of that heavy skew, I don't know how much of deviation comes from longer times, and those are what's ultimately important to me. I'd feel much better seeing the pause times directly, either in a graph or as rows in a table. See also: https://en.wikipedia.org/wiki/Skewness

To_be_clear, I'm not asking for a skew metric - most people won't know or care what that is. I'm asking for a clearer picture of stop events that have longer durations. My pointing at skew is just my way of saying "stddev is nice-to-have, but not actually helpful for this specific concern". read more..

2021-08-03

need Chinese version!!!

2021-08-02

jj